B-cell epitope peptide immunotherapy alleviates chitin-binding protein-induced type 2 airway inflammation in a Blomia tropicalis-murine model

Abstract

Background

Peptide immunotherapy (PIT) offers a safe and effective treatment with minimal side effects. This study aims to identify B-cell epitopes of a novel allergen from Blomia tropicalis (B. tropicalis), specifically the Chitin-binding domain type 2 (ChtBD2) protein, and evaluate the therapeutic effects of peptide treatment in a murine model.

Methods

Using Alphafold2, the 3D structure of ChtBD2 was constructed. AI-based and traditional computational tools predicted the predominant B-cell epitopes. Twelve synthesized peptides were assessed for allergenicity and immunogenicity. A murine model of B. tropicalis-induced allergic airway inflammation mimicking human atopic asthma was developed and analyzed.

Results

Predominant B-cell epitopes of ChtBD2 were identified as promising IgE-binding domains. Peptide 1 (PT1: 1–15) showed significant IgE-binding activity and the highest inhibition rate in competitive IgE-binding assays. PT1 upregulated IL-4, IL-13, and CD63 in B. tropicalis-sensitized patients’ PBMCs and basophils, respectively. Notably, IT groups showed reduced lung cellular infiltration and type 2 cytokine expression in BALF. Specific IgE levels were reduced, with a decline in the IgG₁/IgG_2a ratio.

Conclusions

This study represents the first AI-facilitated development of a B-cell epitope-based ChtBD2 PIT, showing promise as an immunotherapy for B. tropicalis-allergic patients with reduced allergenicity and high immunogenicity in inducing IgG-blocking antibodies.

Clinical trial

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12931-025-03207-8.

Introduction

Allergic asthma is one of the most common chronic airway inflammatory diseases affecting both adults and children globally. In China, the prevalence of asthma among individuals over the age of 20 is 4.26%, totaling 45.7 million people [1]. In patients with allergic asthma, allergen-specific IgE present in the serum can induce local and systemic type 2 inflammatory responses. Although conventional pharmacotherapy is considered the primary treatment, it lacks long-term efficacy, and frequent use may lead to drug resistance and potential side effects. Allergen-specific immunotherapy (AIT) is currently the only treatment that can potentially alter the natural course of allergic diseases, offering long-term efficacy and preventing disease progression [2]. Despite the recognition of AIT’s efficacy by the WHO, the ability of allergen-specific IgE molecules to cross-link mast cells and/or basophils presents significant challenges, as approximately 5–35% of patients undergoing AIT may experience adverse systemic reactions [3]. This issue arises primarily because the allergen vaccines used in AIT are often crude extracts of natural allergens, which exhibit significant variability in quality [4’6], lack critical allergens, and show substantial differences in IgE reactivity [7]. Variations in the quality and potency of allergen extracts raise significant concerns from the perspective of safety and compliance with long, cumbersome AIT regimens [8].

Epitope prediction is a crucial branch of immunoinformatics with significant implications for developing epitope-based vaccines [9]. By using purified recombinant allergens in diagnostics, specific allergens can be identified, while artificial intelligence algorithms can predict protein structures, B-cell epitopes [10, 11], and MHC-II epitopes [12, 13], enabling precision immunotherapy [13]. In recent years, using peptides derived from allergen protein epitopes for desensitization has been considered a promising approach to enhancing the safety of AIT. Common T-cell and B-cell epitope peptides, typically comprising several or dozens of amino acid residues, reduce apparent side effects of immunotherapy due to their lack of the size and necessary three-dimensional structure to cross-link IgE on effector cells. However, natural recombinant allergens still retain IgE reactivity and T-cell epitopes, potentially causing immediate and delayed adverse reactions [14]. Compared to T-cell epitope peptides, B-cell epitope peptides, which are selected based on IgE binding sites and usually consist of short peptides containing a single B-cell epitope, do not cross-link IgE to elicit allergic reactions and thus exhibit lower allergenicity. Currently, the application of B-cell epitopes in AIT focuses mainly on pollen immunotherapy. For instance, B-cell vaccines synthesized from major pollen allergens such as Phl p 1 from Timothy grass pollen, Bet v 1 from birch pollen, and Keyhole Limpet hemocyanin (KLH) have been developed [15, 16]. These B-cell epitope peptides can significantly increase allergen-specific IgG antibodies, thereby inhibiting specific IgE reactivity and basophil activation by natural-like allergens, protecting against allergic inflammation [17]. Furthermore, lacking allergen-specific T-cell epitopes, these peptides do not activate allergen-specific T-cell responses, reducing the production of related inflammatory cytokines and alleviating allergy symptoms.

In the past decade, the prevalence of allergic diseases caused by Blomia tropicalis has been on the rise, potentially altering and influencing global IgE reactivity to mites [18’20]. B. tropicalis is widely recognized as a major source of indoor air allergens, exhibiting heterogeneity in its allergenic components, with varying clinical relevance. The high prevalence of B. tropicalis sensitization necessitates the development of effective treatments with significant efficacy.

In this study, we applied AI strategies to identify B-cell epitopes of the newly discovered potential major allergen with a type 2 chitin-binding domain (ChtBD2) found in B. tropicalis. We characterized the allergenicity and immunogenicity of these B-cell epitope peptides using enzyme-linked immunosorbent assay (ELISA), competitive ELISA inhibition assay, basophil activation test, and cytokine expression analysis in peripheral blood mononuclear cells (PBMCs). Additionally, we established B. tropicalis-sensitized murine models of type 2 allergic asthma to evaluate the immunotherapeutic efficacy of the predicted B-cell epitope peptides. We propose that the AI-driven design of B-cell epitope peptides represents a promising strategy and vaccine candidate for the immunotherapy of B. tropicalis-allergic patients (Fig. 1).

Fig. 1.

Graphical abstract. In this study, both recombinant ChtBD2 and its B-cell derived peptide 1 (PT1) displayed similar efficacy in reducing inflammation by downregulating type 2 cytokines, as well as cell infiltration and mucus production in the lungs of murine model of Blomia tropicalis induced allergic airway inflammation mimicking human atopic asthma.

Methods

Construction and quality assessment of ChtBD2 tertiary structure

To develop peptide vaccines, we used the amino acid sequence of the chitin-binding domain type 2 (ChtBD2) protein, recently identified from Blomia tropicalis (GenBank ID: KAJ6215440.1). This sequence was submitted to AlphaFold2 to create a three-dimensional (3D) model [21]. We assessed the model’s quality using the SAVES v6.0 server (https://saves.mbi.ucla.edu/), including tools like PROCHECK [22] and ERRAT [23]. The structure was then visualized using PyMOL software (https://pymol.org/2/).

Prediction of T-cell epitopes

We used TepiTool to identify regions in the protein that could trigger a T-cell response (http://tools.iedb.org/tepitool/) [24]. The analysis included the 26 most common human class II alleles from HLA-DR, HLA-DQ, and HLA-DP, focusing on sequences with a rank of 10 or less. T-cell epitopes were chosen from regions that appeared in two or more of these alleles.

Prediction of B-cell epitopes and allergenicity of synthesized B-cell epitopes

We predicted linear B-cell epitopes using both AI-based and traditional methods. The 3D structure of ChtBD2 was analyzed with the ElliPro server (http://tools.iedb.org/ellipro/) [25]to find B-cell epitopes, using a score threshold of > 0.5. Additional predictions were made with Bepipred 2.0 (http://tools.iedb.org/bcell/) [26] and GraphBepi (http://bio-web1.nscc-gz.cn/app/graphbepi), with a threshold of 0.5 in Bepipred 2.0. The final B-cell epitopes were selected from regions identified by at least two methods.

We also predicted conformational B-cell epitopes, which are important for allergenicity in inhalant allergens, using the SEMA AI (http://sema.airi.net) [27, 28]method with a threshold of 0.6. All results were combined, and the selected B-cell epitope peptides were synthesized by Taopu-biotech company (Nanjing, China).

Peptide synthesis

Peptides were synthesized via Fmoc-based solid-phase peptide synthesis (SPPS) on Rink Amide MBHA resin (0.5 mmol/g). Resin swelling (DMF, 30 min) was followed by iterative Fmoc deprotection (20% piperidine/DMF), amino acid coupling (HBTU/HOBt/DIEA, 4 equiv, 1 h), and DMF washes. Challenging sequences utilized microwave assistance (50 °C, 20 min) or anti-aggregation agents. Peptides were cleaved with TFA/H2O/TIS/EDT (95:2.5:1.25:1.25, 2 h), precipitated in cold ether, and lyophilized. Purification by RP-HPLC (C18, 0.1% TFA/ACN gradient) yielded > 95% pure peptides (HPLC, MALDI-TOF). Peptides were stored at − 80 °C.

Subjects and Sera collection

This study protocol was approved by the Institutional Review Board (IRB) of the First Affiliated Hospital of Guangzhou Medical University (GYFYY-2017-18). We recruited 107 patients sensitized to Blomia tropicalis from the Allergy Clinic, confirmed by sIgE levels above 0.35 kU/L using the ImmunoCAP system. Among them, 74 were female and 33 were male, with a median age of 18 years (ranging from 3 to 51 years) and an average sIgE level of 7.12 kU/L. We also recruited ten non-allergic participants as controls, as previously described [29]. Informed consent was obtained from all participants or their guardians. Further details are provided in Additional File S1.

IgE reactivity analyzed by indirect ELISA and competitive ELISA Inhibition test

We coated 96-well plates with twelve purified ChtBD2-derived B-cell epitope peptides at a concentration of 10 µg/mL and incubated them overnight at 4 °C. The plates were then blocked with 1% skim milk powder at 37 °C for 2 h. Fourteen IgE-positive sera from the B. tropicalis-sensitized participants were diluted 1:2 in PBST and added to the plates, followed by a 1-hour incubation at 37 °C. After washing, we added goat anti-human IgE secondary antibodies coupled with HRP and incubated for another hour at 37 °C. Optical density (OD) was measured at 450 nm. We set the positivity threshold as the mean plus three standard deviations of the OD values from the ten non-allergic controls.

We used a competitive ELISA to evaluate the IgE-binding strength of twelve peptides. Sera from ten sensitized patients were pre-incubated with each peptide at a concentration of 100 µg/mL at 37 °C for 2 h. The pre-treated serum was then added to 96-well plates coated with the corresponding peptides. The inhibition rate was calculated as follows: Inhibition rate (%) = (OD450 uninhibited − OD450 inhibited) / (OD450 uninhibited OD450 control) × 100%.

Cytokine expression in PBMCs

We collected fresh blood from five individuals who were sensitized to Blomia tropicalis and allergic to ChtBD2. PBMCs were isolated and cultured in 96-well plates at a density of 1.0 × 10⁶ cells per well. The cells were stimulated with five candidate peptides (100 µg/mL) for 72 h after culturing for 24 h at 37 °C in a humidified atmosphere with 5% CO2. Culture medium was used as a negative control. We measured the levels of cytokines IL-4 and IL-13 in the supernatants using ELISA kits (Cusabio, China).

Basophil activation test (BAT)

We collected 5 mL of fresh heparinized blood from five patients sensitized to Blomia tropicalis who tested positive for ChtBD2-specific IgE. Anti-FcεRI antibodies (PC1) and PBS were used as positive and negative controls, respectively. Five candidate peptides (PT1-5) were diluted in PBS to a concentration of 100 µg/µL and incubated with the whole blood samples at 37 °C for 15 min. The reaction was stopped with 2.0 mM EDTA, followed by staining with a mixture of antibodies: FITC-conjugated anti-CD45, PE-conjugated anti-CD123, PerCP-conjugated anti-HLA-DR, and Alexa Fluor 647-conjugated anti-CD63. The samples were then lysed, washed, and analyzed using FACS Canto II, measuring basophil activation by the expression of CD63. Activated basophils were identified as SSC^low and CD63⁺ events. Stimulation index (SI) was calculated as the percentage of allergen/peptide-activated basophils/percentage of spontaneously activated basophils (PBS) and SI > 2 was defined as a positive result.

Establishment and evaluation of the murine model of B. tropicalis-induced allergic airway inflammation mimicking human (atopic) asthma

We used female Balb/c mice, aged 6–8 weeks and weighing 18–20 g, from the Guangdong Medical Laboratory Animal Center. The mice were kept under standard conditions. The experimental procedures were approved by the First Affiliated Hospital of Guangzhou Medical University (No. 20240020). Mice are sensitized with i.p. injections of PBS or ChtBD2-protein (25 µg) on Days 0, 7, and 14. IT is administered on Days 21–29, and mice are challenged with i.n. injections of PBS or ChtBD2-protein (50 µg) on Days 36–40. Mice are euthanized on Day 41. Non-IT group did not receive any immunotherapy treatment and was regarded as a positive control. Airway responsiveness was measured 24 h after the final challenge using the Buxco RC system. We assessed respiratory mechanics before challenging the mice with increasing concentrations of nebulized methacholine (6.25, 12.5, 25, and 50 mg/mL). Additional analyses included bronchoalveolar lavage fluid (BALF), lung tissue histology, serology, and cytokine expression (IL-4, IL-5, IL-13, IL-10, and IFN-γ).

Statistical analysis

Quantitative variables were described as mean and standard deviation (mean ± S.D.), while qualitative variables were expressed as frequency or percentage. Different data were analyzed by one-way ANOVA or Dunnett’s test with SPSS v. 22 software (IBM Corp., USA). Charts were plotted by GraphPad Prism v. 7 (San Diego, USA) and R v. 4.2.1(Core Team 2022). A p‐value < 0.05 was regarded as statistically significant. Significance displayed in figures as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001 and ****, p < 0.0001.

Results

Expression and purification of ChtBD2

The ChtBD2 protein, tagged with a 6-His marker at its N-terminus, was expressed using a modified pET-28a plasmid. The protein was then purified through affinity chromatography using Ni-NTA agarose resin. After cleavage by thrombin, the protein’s purity and molecular weight was evaluated.

3D modeling and evaluation of ChtBD2 structure

The 3D structure of ChtBD2 was constructed using AlphaFold2, based on its amino acid sequence. The model revealed that ChtBD2 is composed of 104 amino acid residues, featuring a chitin-binding domain from residues 5 to 57 (Fig. 2A). The SDS-PAGE gel image showed that the molecular weight of ChtBD2 was nearly 16 kDa (Fig. 2B). In the structural model, different colors represent different secondary structures, with the model showing pink loops and five purple β-sheets (Fig. 2C). The overall quality factor, as evaluated by the ERRAT program, was 23.077 (Fig. 2D). PROCHECK analysis showed that 47.7% of the amino acid residues were in core regions, 31.4% in allowed regions, 19.8% in generously allowed regions, and 1.2% in disallowed regions. Only one residue (GLU-103) was in the disallowed regions of the Ramachandran plot (Fig. 2E).

Fig. 2.

Identification of dominant epitopes of ChtBD2 and its 3D modeling and evaluation. (A) Distribution of B-cell epitopes in the amino acid sequence of ChtBD2. Five candidate dominant linear B-cell epitopes are shaded. (B) Purified ChtBD2 protein was separated by 15% SDS-PAGE and then stained with Coomassie Brilliant Blue. (C) Alphafold2 constructed a 3D structure of ChtBD2 protein. The secondary structure of the ChtBD2 protein is mainly composed of pink loops and 5 purple β-sheets. (D) ERRAT program evaluation. (E) PROCHECK analysis shows the distribution of the proportion of amino acid residues located in the core region, allowed region, generous allowed region and unallowed region. (F) Distribution of five candidate dominant linear B-cell epitopes in 3D structure of ChtBD2

Computational prediction of T-cell epitopes of ChtBD2

We used TepiTool to predict which regions of the ChtBD2 protein could bind to MHC class II molecules, identifying the following regions as potential T-cell epitopes: residues 11–25, 30–44, 38–52, 49–63, 55–69, 66–80, 83–97, and 88–102. A lower percentile rank indicates a better binding affinity. Detailed information on each epitope and its corresponding binding allele is provided in Supplementary Table S1.

Prediction of B-cell linear and discontinuous epitopes of ChtBD2

We predicted linear B-cell epitopes of ChtBD2 using three different servers. According to ElliPro, three linear epitopes were predicted at residues 1–12, 64–86, and 91–104 (Supplementary Table S2). Bepipred 2.0 identified five epitopes: residues 5–13, 15–16, 18–23, 43–52, and 62–102 (Supplementary Table S3). Graphbepi predicted three epitopes: residues 28–36, 49–52, and 94–100 (Supplementary Table S4). Additionally, the SEMA server predicted 44 residues as conformational epitopes, many of which overlapped with the linear epitopes (Supplementary Table S5). We combined all the predicted linear and conformational epitopes to select the final sequence (Table 1). These B-cell epitope fragments, excluding T-cell epitopes, were highlighted in both the 1D and 3D structures (Fig. 2F). The final selected fragments were Fragment 1 (1–12), Fragment 2 (18–23), Fragment 3 (28–36), Fragment 4 (46–52), and Fragment 5 (97–104).

Table 1.

Detailed information of predicted linear B-cell epitopes of ChtBD2

Sever	Residues
Ellipro	1–12, 64–86, 91–104
BepiPred 2.0	5–13, 15–16, 18–23, 43–52, 62–102
Graphbepi	28–36, 49–52, 94–100

IgE-binding affinity of 12 peptides and IgE competitive Inhibition test

Based on the predictions, we synthesized twelve overlapping peptides, ensuring seven identical amino acid residues between adjacent peptides. The potential of these peptides to serve as linear B-cell epitopes was further evaluated using machine learning combined with position weight matrix (PWM) analysis. Results showed that PT1 (¹MYYEDFHCPPEDGFY¹⁵), PT3 (¹⁷CPENPHKFYECIEGM³¹), PT4 (²⁵YECIEGMEYLFTCPA³⁹), PT5 (³³YLFTCPANLLFNESK⁴⁷) and PT12 (⁹⁰TIDSRHSTRGRHHED¹⁰⁴) were identified as potential B-cell epitopes by at least two independent computational methods. PT8 (⁵⁷IVVRTTKAPATPEHG⁷¹) and PT9 (⁶⁵PATPEHGTRPRHTEP⁷⁹) met the threshold for one method, suggesting that the strongest IgE-binding regions are likely in the earlier part of the sequence (Table 2).

Table 2.

Twelve synthetic peptides of linear B cell epitopes of ChtBD2

PT. No	Peptide sequence	ML.Predict	PWM.type.I	PWM.type.II	Peptide range
1	MYYEDFHCPPEDGFY	0.81	31.99	0.01	1–15
2	PPEDGFYRCPENPHK	0.22	0.80	0.16	9–23
3	CPENPHKFYECIEGM	0.60	2.72	0.19	17–31
4	YECIEGMEYLFTCPA	0.80	3.74	0.02	25–39
5	YLFTCPANLLFNESK	0.64	1.29	0.06	33–47
6	LLFNESKQKCDYVTG	0.31	0.81	0.41	41–55
7	KCDYVTGPIVVRTTK	0.13	0.77	0.14	49–63
8	IVVRTTKAPATPEHG	0.04	0.38	16.3	57–71
9	PATPEHGTRPRHTEP	0.00	0.59	3.56	65–79
10	RPRHTEPNTKPDQYE	0.01	0.19	0.49	73–87
11	TKPDQYESTTIDSRH	0.06	0.74	0.87	81–95
12	TIDSRHSTRGRHHED	0.17	4.21	2.35	90–104

We then evaluated the IgE-binding reactivity of these twelve peptides using an indirect ELISA with sera from 14 ChtBD2-positive patients and a pooled serum from 5 negative controls. The results indicated that the most reactive regions were at the beginning and end of the sequence. The peptides with the highest sensitization rates were PT1, PT12, PT11, PT3, PT10, and PT9, aligning with the in-silico predictions (Fig. 3A). Additionally, a competitive ELISA was performed using sera from 10 ChtBD2-allergic patients. PT1 showed the highest inhibition rate (just under 30%), followed by PT6, PT9, PT8, and PT12. A higher inhibition rate indicates a stronger IgE-binding affinity (Fig. 3B).

Fig. 3.

Evaluation of the IgE-binding affinity of twelve synthetic peptides of ChtBD2. (A) The IgE binding reactivity of 12 peptides was evaluated by indirect ELISA. The color depth of the heatmap represents the binding ability of twelve peptides to IgE, and the bar graph is the average binding affinity of all samples. (B) Competitive ELISA assays were performed to detect the IgE reactivity of the synthesized B-cell epitope fragments

Cytokine responses of PBMCs

To assess the immunogenicity of ChtBD2 B-cell epitopes, we cultured PBMCs from five Blomia tropicalis-allergic patients with five candidate peptides. The levels of type 2 cytokines were then measured by ELISA. The results showed that all five candidate peptides could induce a significant increase in type 2 cytokine production. Notably, PT1 induced higher levels of IL-4 (112.18 ± 15.01 pg/mL) and IL-13 (110.32 ± 8.22 pg/mL) compared to the control group (29.98 ± 3.59 pg/mL for IL-4 and 8.89 ± 2.54 pg/mL for IL-13, p < 0.001) (Fig. 4).

Fig. 4.

Cytokine responses of PBMCs. Comparison of PBMC cytokine expression after stimulating with five dominant B-cell linear peptides. Stimulated with PT1-5 respectively (final concentration of 100 µg/mL), supernatant fluids were collected for cytokine (IL-4 and IL-13) ELISA. Values were analyzed by the one-way ANOVA test and Dunnett’s test, as post hoc analysis. A p-value < 0.05 is regarded as statistically significant. Statistical significance in the figures is indicated as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001 and ****, p < 0.0001

Amelioration of pulmonary inflammation in mice by B-cell peptide immunotherapy (PIT)

Schematic representation of immunization protocol (Fig. 5A). Compared with the Non-IT group, ARI values of ChtBD2 and PT1 immunized mice started to show significant differences from a Mch dose of 12.5 mg/mL, indicating an improving airway responsiveness (Fig. 5B).

Fig. 5.

Assessment of pulmonary inflammation in a murine model of Blomia tropicalis-induced atopic asthma. (A) Schematic representation of immunization protocol followed for evaluation of the immunotherapeutic potential of ChtBD2-derived B-cell epitope peptides. (B) Airway responsiveness is measured by the Buxco RC system after 24 h at the last challenge, presenting as the airway resistance index. (C) Representative images of stained hematoxylin-eosin (H&E) and (D) periodic acid-Schiff (PAS) lung sections. An inflammation and PAS score are calculated and presented as mean + SEM (n = 5 mice per group) respectively after observation of 10 consecutive fields per slide. (E) Representative images of stained H&E of nucleated cells in bronchoalveolar lavage fluid (BALF). (F) Nucleated cell counts in BALF. (G) Proportions of major inflammatory cell populations in BALF. MAC, macrophages; EOS, eosinophils; NEU, neutrophils; LYM, lymphocytes. Between-group comparisons are performed using the one-way ANOVA test and Dunnett’s test, as post hoc analysis. A p-value < 0.05 is regarded as statistically significant. Statistical significance in the figures is indicated as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001 and ****, p < 0.0001

Lung histological examination by Hematoxylin and eosin (H&E) staining revealed the severity of cellular infiltration and disruption of alveolar structure in different groups. As is illustrated in sections, the swelling and thickening of lung bronchiole, exfoliation of epithelial cells and infiltration of inflammatory cells were observed obviously in the Non-IT mice group (Fig. 5C). In contrast, the ChtBD2-IT and PT1-IT groups were able to alleviate lung inflammation. The maximum reduction in inflammation score evaluated by three experienced medical practitioners majoring in clinical pathology was observed in the ChtBD2-IT group (3-fold change with respect to Non-IT) followed by PT1-IT mice (2-fold reduction with respect to Non-IT). Besides, positive reactions were revealed by the periodic acid-Schiff (PAS) lung sections in allergen-induced mice groups, indicating hyperplasia of goblet cells and excessive secretion of mucus compared with the PBS group, leading to the accumulation of mucus in the airway cavity (Fig. 5D). Correlation analysis between IL-13 levels in BALF and PAS score revealed that IL-13 level was positively correlated with PAS-positive goblet cell score (R² = 0.983). Goblet cells count in PAS-stained lung histological slides demonstrated a significant reduction in IT-administered groups in comparison to Non-IT mice groups. Both ChtBD2-IT and PT1-IT mice groups showed remarkable improvement in airway mucus production.

Nucleated cell counts in BALF were significantly elevated in Non-IT, ChtBD2-IT, and PT1-IT groups compared to Control group, with pronounced inflammatory infiltration, confirming successful murine models of allergic asthma induction. Compared to Non-IT group, ChtBD2-IT and PT1-IT group reduced total BALF cells by 3.7-fold and 2.6-fold (p < 0.05), respectively, with attenuated inflammation (Fig. 5E, F). Cell differential counting showed that eosinophils (EOS accounted for 60% on average) and neutrophils (NEU accounted for 12% on average) significantly increased in Non-IT group, and macrophages (MAC) accounted for 20% on average. Post-intervention, inflammatory cells (EOS and NEU) in BALF decreased significantly in ChtBD2-IT and PT1-IT group, while the average cell proportion rate of MAC and lymphocytes (LYM) increased (Fig. 5G).

Modulation of serum immunoglobulins (IGs) profiles and type 2 immune response suppression through B-cell peptide immunotherapy

The profile of immunoglobulin isotypes is heavily influenced by the type 1/type 2 balance of the immune response. Allergic asthma is a typical type 2 immune reaction closely related to one’s immunity homeostasis. In this study, we detected 8 types of serum IG levels, including total IgE (tIgE), ChtBD2-specific IgE (sIgE), IgA, IgG₁, IgG_2a, IgG_2b, IgG₃ and IgM, to evaluate the peptides immunotherapy efficacy in different mice groups (Fig. 6A).

Fig. 6.

Regulation of serum immunoglobulins (IGs) profiles via ChtBD2-derived B-cell peptide immunotherapy in Blomia tropicalis-induced allergic asthma murine models. (A) Eight serum IGs, including total IgE (tIgE), ChtBD2-specific IgE (sIgE), IgA, IgG₁, IgG_2a, IgG_2b, IgG₃, and IgM, were detected and evaluated the immunotherapeutic effects of peptides in mice. Data are reported as individual OD. (B) Evaluation of type 1- (IL-10, IFN-γ) and type 2- (IL-4, IL-5, IL-13) cytokines expression profile in BALF. Between-group comparisons are performed using the one-way ANOVA test and Dunnett’s test, as post hoc analysis. A p-value < 0.05 is regarded as statistically significant. Statistical significance in the figures is indicated as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001 and ****, p < 0.0001

Firstly, serum IgE is the principal mediator of allergic reactions and therefore the effect of immunotherapy was examined on ChtBD2-specific IgE. An obvious decline in ChtBD2 allergen-specific IgE antibody was observed in ChtBD2-IT and B-cell peptide PT1-IT groups. Compared to the Non-IT group, about three times the reduction in ChtBD2-specific IgE levels was observed in the ChtBD2-IT group followed by PT1-IT (3.3- and 2.8-fold reduction, respectively). Moreover, tIgE levels of ChtBD2-IT and PT1-IT groups shared the same tendency with ChtBD2-specific IgE, displaying about twice the decline compared with the Non-IT mice group (2.0- and 1.9-fold reduction, respectively). Analysis of the ratio of IgG₁ to IgG_2a antibodies revealed maximum downregulated levels in the PT1-IT group (5.5-fold) followed by the ChtBD2-IT group (4.8 times) compared with the Non-IT group. By contrast, IL-4 has powerful effects in promoting switching to the expression of IgG₁ and IgE but markedly inhibits IgG₃ and IgG_2b. Compared with the PBS mice group, IgG_2b and IgG₃ levels of IT mice groups were increased to some extent, indicating the IGs switching may be induced by type 1 responses. A decline of IgA level in IT mice groups was observed in comparison with PBS and Non-IT groups. While compared with the Non-IT group, results showed that IgM antibodies revealed downregulated levels in both ChtBD2-IT and PT1-IT mice groups.

Administration of ChtBD2 and PT1 to AIT group mice respectively (ChtBD2-IT and PT1-IT groups) was able to reduce the levels of type 2 cytokines (IL-4, IL-5 and IL-13) and increase the type 1 cytokines (IL-10 and IFN-γ) in BALF significantly demonstrating an overall shift from type 2-dependent pro-inflammatory microenvironment (with more than 2-fold reduction in IL-4 levels and more than 1.5 times reduction in IL-13 cytokine levels) (Fig. 6B). Notably, even though statistically significant differences of IL-5 levels were observed between the Control group and all intervention groups, the differences between Non-IT group and IT groups (ChtBD2-IT and PT1-IT) did not reach statistical significance. Detailed information please refer to Supplementary Table S6.

Basophil CD63 upregulation mediated by ChtBD2-derived B-cell epitopes

Basophil activation test was performed in a whole blood pool from five B. tropicalis-sensitized and ChtBD2-positive-reaction blood donors. Five candidate B-cell peptides co-cultured with whole blood and subsequently detected the expression of CD63. In comparison with the PBS group, results show that PT3 elicited the highest CD63 upregulation, accounting for an average proportion (%) of 30.3 in basophils, followed by PT1, PT5, PT4 and PT2, with an average CD63-positive cell rate of 29.1, 26.7, 22.3 and 11.1 respectively (Fig. 7A). Stimulation index (SI) showed a similar tendency with CD63-positive cell rate (Fig. 7B). The gating strategy for basophils and representative flow cytometry plots are shown in Supplementary Figure S1.

Fig. 7.

Basophil CD63 upregulation mediated by ChtBD2-derived B-cell epitopes. (A) Percentages of CD63⁺ basophils after stimulating by different epitopes and (B) corresponding stimulation index (SI). The basophils of Blomia tropicalis-allergic patients were stimulated with PBS (negative control), anti-FcƐRI (positive control) or PT1-5 (with a final concentration of 100 µg/µL), respectively. SI was calculated as the percentage of allergen/peptide-activated basophils/percentage of spontaneously activated basophils (PBS) and SI > 2 was defined as a positive result. Between-group comparisons are performed using the one-way ANOVA test and Dunnett’s test, as post hoc analysis. A p-value < 0.05 is regarded as statistically significant. Statistical significance in the figures is indicated as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001 and ****, p < 0.0001

Discussion

Allergen-specific immunotherapy (AIT) has been successfully implemented in clinical treatments for various allergens such as house dust mites and Artemisia pollen [30, 31]. It is currently considered the only method that can alter the course of the disease to achieve a sustained therapeutic effect. Pharmacotherapy may still be required in certain situations, even during AIT, and continues to offer therapeutic benefits. However, the AIT used in clinics has unavoidable shortcomings, such as inconsistent purity and standardization of allergen preparations, long treatment courses (3–5 years), a high risk of adverse reactions (such as IgE cross-linking-induced acute allergic reactions or delayed allergic reactions through effector T-cell activation), and poor patient compliance [32]. Over the past decades, researchers have been striving to improve AIT techniques to address these issues. Peptide immunotherapy (PIT) is considered a promising candidate for allergy treatment. These small molecular peptides, formed by enzymatic degradation of allergens, are recognized by the immune system’s T or B cells and are referred to as T-cell epitopes or B-cell epitopes. PIT uses these epitopes to prepare vaccines, inducing tolerance to the allergen sources in treated patients. As PIT peptides are linear structures composed of several to dozens of amino acids, they can induce effector T-cell “anergy”, inhibit IgE synthesis, promote IgG₄ secretion, and induce tolerance to allergens. Due to their short length and lack of three-dimensional structure, the likelihood of being recognized by IgE and triggering downstream inflammatory responses is low. Even though PIT is safer than traditional AIT, some adverse events, mainly late-phase reactions such as rhinitis, skin reactions, conjunctivitis, and flushing, have been observed in some clinical trials [33’36].

Identifying epitopes within allergens is of significant interest for understanding disease etiology, immune monitoring, developing diagnostic assays, and designing epitope-based vaccines. However, epitope identification is costly and time-consuming, requiring experimental screening of many potential candidates. Fortunately, researchers have developed numerous computational prediction methods to reduce the list of potential epitope candidates for experimental testing. AI-designed T/B cell epitope-based peptides represent a major advancement in recombinant allergen derivatives. AI tools, widely used in vaccine research, facilitate rational design and enhance efficiency [37]. Most B-cell epitopes (approximately 90%) are conformational, but predicting conformational epitopes is unreliable due to unclear structural contributions to antigen-antibody binding. Similarly, for inhalation allergens, conformational epitopes also appear to be major targets for IgE binding [38]. Studies show that short residues of B-cell epitopes (7–12 amino acids) cannot form epitope-complement complexes [39]. To improve computational predictions of immunodominant B-cell epitopes, it is recommended to exclude short epitopes, analyze candidate epitopes based on protein 3D structure characteristics (such as hydrophilicity, flexibility, surface exposure, and solvent accessibility), use multiple prediction methods based on different algorithms, and conduct molecular interaction analysis with online B-cell epitope prediction tools (e.g., molecular docking and dynamics simulations).

B-cell epitopes on allergen proteins are directly recognized by B cells without special processing. Accurate identification of B-cell epitopes is fundamental for developing antibody therapies, peptide vaccines, and immunodiagnostic tools. Unlike recombinant allergens, single B-cell epitopes do not cause IgE cross-linking, resulting in lower allergenicity. Furthermore, these vaccines do not contain allergen-specific T-cell epitopes, reducing related inflammatory cytokine production and alleviating allergic symptoms. The B-cell epitope-based recombinant vaccine BM32, combining major B-cell epitopes from Timothy grass allergens Phl p 1, 2, 5, and 6 with the hepatitis B virus surface protein, has shown protective immune responses to both allergens and viruses, inducing beneficial clinical effects in multiple trials [36, 40, 41]. Studies indicate that B-cell epitope vaccines reduce patch-test-positive reactions caused by allergen-specific T-cell epitopes and do not induce IgE-mediated allergic skin reactions [42, 43].

Although five candidate peptides showed varying degrees of binding activity with ChtBD2-sIgE, we considered all of them for desensitization treatment in allergic asthma mice. B-cell peptides’ short sequences do not induce IgE cross-linking while maintaining immunogenicity and reducing allergenicity. Results indicated that B-cell peptide immunotherapy effectively alleviated type 2 asthma induced by ChtBD2 protein, demonstrated by lung histopathology, inflammation scores, and cytokine levels in BALF. Allergic individuals with specific IgE in serum tend to produce T-helper cell type 2 cytokines [44]. IL-4, IL-5, and IL-13 are considered type 2 cytokines, while IL-10 and IFN-γ have protective roles in type 2 asthma [45]. The duality of IL-10 in the immune response may vary depending on disease endomorphism or microenvironmental cues [46]. Desensitization with B-cell epitopes significantly downregulated type 2 cytokines IL-4 and IL-13 in BALF while promoting type 1 cytokines IL-10 and IFN-γ secretion, potentially aiding in T-cell tolerance induction due to IL-10’s immunosuppressive function [47, 48]. Post-desensitization treatment showed significant reductions in specific and total IgE, along with a decreased IgG₁ to IgG_2a ratio in mouse serum. Lowering allergen-specific IgE levels and producing blocking IgG_2a antibodies can relieve type 2 inflammation through AIT. Studies indicate that IL-4-dominated type 2 responses promote IgE class switching and higher IgG₁ to IgG_2a ratios, while type 1 responses driven by IL-12 and IFN-γ lead to higher IgG_2a levels and lower IgG₁ and IgE ratios [49]. In mice, IgG₃ switching is also associated with type 1 response induction, with high Fcγ receptor affinity and complement fixation [50]. Possible explanations for the lack of significant increase in IgA antibody levels in mice following PIT. First, the lack of significant IgA elevation post-AIT in murine models may stem from mucosal-systemic compartmentalization, as IgA primarily mediates localized immunity in respiratory/gastrointestinal tracts rather than systemic responses [51, 52]. Intraperitoneal administration preferentially activates splenic/peritoneal lymph nodes, driving IgG over mucosal IgA production, while sublingual routes favor IgA responses [53]. Systemic immunization may fail to engage mucosal lymphoid tissues, limiting detectable serum IgA elevation [54]. Second, antigen structural constraints also play a role, as short peptides often lack conformational epitopes required for IgA class-switching, presenting linear epitopes insufficient to activate IgA⁺ plasma cells, while multivalent antigen-BCR crosslinking, critical for IgA differentiation, is impaired by monomeric peptides [51]. Third, cytokine imbalances, such as TGF-β deficiency in type 1/type 17-polarized microenvironments, disrupt IgA class-switch recombination, and concurrent type 2 cytokines promote IgE/IgG₁ over IgA [55, 56]. Lastly, methodological limitations further complicate detection, as mucosal IgA responses require longitudinal sampling due to delayed kinetics [57], serum IgA poorly correlates with mucosal sIgA in BALF/intestinal secretions, and unvalidated murine IgA kits may exhibit cross-reactivity or inadequate detection thresholds.

Continuous discovery and design of new allergens and allergen derivatives enhance understanding of allergenicity, aiding new diagnostic and immunotherapy methods. Despite promising results indicating AI’s role in identifying dominant epitopes, facilitating future allergen immunotherapy vaccine development, and reducing costs and time, several limitations should be noted. Only the peptide with the strongest binding activity to ChtBD2-sIgE was used for in vivo desensitization, lacking results for the other four candidate peptides. Future studies should evaluate the desensitization effects of these peptides to improve vaccine development. Meanwhile, to better evaluate the immune status of mice, if possible, analysis of T regulatory cells (CD4⁺Foxp3⁺ cells) should be carried out in the splenocyte population. If an increase in the population of T-regulatory cells is observed in mice in the immunotherapy group, this implies that a systemic adaptive regulatory mechanism has been induced.

Given the limitations of allergen extract-based technologies and advances in allergen research, molecular approaches for AIT have been developed. Molecular therapy shows great promise, as successful immunotherapy trials with recombinant allergens and allergen derivatives (e.g., T/B cell peptides or recombinant low-allergenicity allergen derivatives) indicate that molecular therapy methods and alternative administration routes can improve AIT safety and efficacy [37]. Currently, desensitization strategies focus on major allergenic components (IgE binding frequency > 50%). For example, researchers have developed several low-allergenicity allergen derivatives with reduced IgE reactivity, such as Der p 1, Der p 2, and Der p 23, for house dust mites. AI applications have also accelerated clinical trials, reducing vaccine development costs and time [58, 59]. Peptide immunotherapy offers distinct pharmacological benefits, including reduced IgE cross-linking capacity due to the absence of conformational epitopes, which minimize anaphylactic risks during treatment. Additionally, synthetic peptides bypass proteolytic processing requirements, enabling precise targeting of immunodominant epitopes while avoiding non-specific immune activation. The structural simplicity of peptide enhances batch-to-batch reproducibility, a critical factor for clinical translation, whereas full-length allergens exhibit natural variability in post-translational modifications (e.g., glycosylation) that may alter immunogenicity. Mechanistically, Peptides focused epitope presentation likely promotes type 1/Treg-biased responses through selective MHC-II binding, contrasting with the broader type 2 potential of full-length antigens containing auxiliary IgE-binding sites.

Conclusions

Both recombinant ChtBD2 and its B-cell derived peptide 1 displayed similar efficacy in reducing inflammation by downregulating type 2 cytokines, as well as cell infiltration and mucus production in the lungs of murine model of Blomia tropicalis induced allergic airway inflammation mimicking human atopic asthma. Overall, AI-predicted and synthesized dominant B-cell epitope peptides of ChtBD2 represent a promising molecular vaccine candidate for tropical mite allergy immunotherapy, providing a theoretical basis for future tropical mite allergen vaccine development.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Jiale Zhang: Conceptualization, Formal analysis, and Writing- Original draft preparation; Wenting Luo: Data curation, Methodology and Visualization; Yubao Cui: Funding acquisition, Writing- Reviewing and Editing; Baoqing Sun: Funding acquisition, Writing- Reviewing and Editing. All authors reviewed and approved the manuscript.

Funding

This study was supported by the Clinical and Epidemiological Research Project of State Key Laboratory of Respiratory Disease (SKLRD-L-202505), Guangzhou Medical University (GMUCR2024-01009), Guangzhou Science and Technology Foundation (2023A03J0365) and Guangdong Zhong Nanshan Medical Foundation (20240015).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The study was in accordance with the Declaration of Helsinki and approved by the ethics committee of the First Affiliated Hospital of Guangzhou Medical University (GYFYY-2017-18, No. 20240020). All authors consented to participate in this work and approved the final version of the manuscript.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.